Cervical Cancer Prevention

Human papillomavirus (HPV)

Based on solid evidence from observational studies, HPV infection is associated with the development of cervical cancer.

Magnitude of Effect: HPV has been implicated as the primary etiologic infectious agent causing virtually all cases of cervical cancer.

Immunosuppression

Based on solid evidence, being immunosuppressed is associated with an increased risk of cervical cancer.

Sexual activity at an early age and with a greater number of partners

Based on solid evidence, sexual activity at a younger age and an increasing number of sexual partners are both associated with an increased risk of HPV infection and subsequent development of cervical cancer.

Magnitude of Effect: Women who experience first sexual intercourse at age 17 years or younger or women who have had six or more lifetime sexual partners have approximately two to three times the risk of squamous cell carcinoma or adenocarcinoma of the cervix, compared with women aged 21 years or older or who have a single sexual partner.[3]

High parity

Based on solid evidence, high parity is associated with increased risk of cervical cancer in HPV-infected women.

Magnitude of Effect: Among HPV-infected women, those who have had seven or more full-term pregnancies have approximately four times the risk of squamous cell cancer compared with nulliparous women, and HPV-infected women also have two to three times the risk of women who have had one or two full-term pregnancies.[6]

Long-term use of oral contraceptives

Based on solid evidence, long-term use of oral contraceptives is associated with increased risk of cervical cancer in HPV-infected women.

Magnitude of Effect: Among HPV-infected women, those who used oral contraceptives for 5 to 9 years have approximately three times the incidence of invasive cancer, and those who used them for 10 years or longer have approximately four times the risk.[7]

Cigarette smoke exposure

Based on solid evidence, cigarette smoking, both active and passive, is associated with an increased risk of cervical cancer in HPV-infected women.

Magnitude of Effect: Among HPV-infected women, current and former smokers have approximately two to three times the incidence of high-grade cervical intraepithelial neoplasia or invasive cancer. Passive smoking is also associated with increased risk but to a lesser extent.

Diethylstilbestrol (DES) exposure

Based on solid evidence, DES exposure is associated with an increased risk of developing clear cell adenocarcinoma of the cervix.

Magnitude of Effect: About one in 1,000 women exposed to DES in utero will develop a clear cell adenocarcinoma of the cervix.

Sexual abstinence

Based on solid evidence, abstinence from sexual activity is associated with a near-total reduction in the risk of developing cervical cancer.

Magnitude of Effect: Sexual abstinence essentially precludes HPV transmission.

HPV vaccination: benefits

Based on solid evidence, vaccination against HPV-16/HPV-18 is effective in preventing HPV infection in HPV-naive individuals and is associated with a reduced incidence of cervical intraepithelial neoplasia 2 and 3. By extrapolation, these vaccines should also be associated with a reduced incidence of cervical cancer.

Magnitude of Effect: Vaccination against HPV-16 and HPV-18 reduces incident and persistent infections with efficacy of 91.6% (95% confidence interval [CI], 64.5%–98.0%) and 100% (95% CI, 45%–100%), respectively. Efficacy beyond 6 to 8 years is not known.

HPV vaccination: harms

Based on solid evidence, harms of HPV vaccines include injection-site reactions, dizziness and syncope, headache, and fever. Vaccination during pregnancy has not been associated with adverse pregnancy outcomes.[10] Allergic reactions occur rarely.

Use of barrier protection during sexual intercourse: benefits

Based on solid evidence, the use of barrier methods (e.g., condoms) during sexual intercourse is associated with a decreased risk of cervical cancer.

Magnitude of Effect: Total use of barrier protection decreases cervical cancer incidence (relative risk, 0.4; 95% CI, 0.2–0.9).

Use of barrier protection during sexual intercourse: harms

Based on fair evidence, the use of barrier methods during sexual intercourse is associated with few serious harms. Barrier methods can break, potentially resulting in unintended pregnancy. Allergic reactions to the barrier material (e.g., natural latex) can occur.

HPV

Epidemiologic studies to evaluate risk factors for the development of squamous intraepithelial lesions (SIL) and cervical malignancy demonstrate conclusively a sexual mode of transmission of a carcinogen.[4] It is now widely accepted that human papillomavirus (HPV) is the primary etiologic infectious agent that causes virtually all cases of cervical cancer.[5,6] Other sexually transmitted factors, including herpes simplex virus 2 and Chlamydia trachomatis, may play a cocausative role.[7] More than 80 distinct types of HPV have been identified, approximately 30 of which infect the human genital tract. HPV type 16 (HPV-16) and HPV type 18 (HPV-18) are most often associated with invasive disease. Characterization of carcinogenic risk associated with HPV types is an important step in the process of developing a combination HPV vaccine for the prevention of cervical neoplasia. In a population-based study of HPV infection and cervical neoplasia in Costa Rica, 80% of HSIL and invasive lesions were associated with HPV infection by one or more of 13 cancer-associated types.[8] In this study, the risk of about 50% of HSIL and invasive cervical cancer was attributable to HPV-16. HPV-18 was associated with 15% of invasive disease but only 5% of HSIL, suggesting that HPV-18 may have a role in more aggressive cases of cervical malignancy.

Immunosuppression

Most cases of HPV infection are resolved by the host immune system. Immunosuppression leads to persistence of viral infection with a subsequent increased risk of cervical neoplasia. Women with immunosuppression resulting from human immunodeficiency virus (HIV) infection have been studied over the past three decades of the AIDS epidemic. In one North American study, a group of 13,690 HIV-infected women were studied for a median of 5 years. The rate of invasive cervical cancer in the HIV-infected women was 26 cases per 100,000 women, and this was approximately four times greater than an HIV-uninfected control group.[9] HIV-infected women with the lowest CD4 lymphocyte counts were at the highest risk of invasive cancer. Women who are immunosuppressed resulting from organ transplantation are also at risk of invasive cervical cancer, and one meta-analysis found a twofold increased risk.[10]

Sexual activity at an early age and with a greater number of partners

HPV infection has been established as a necessary cause of almost all cases of cervical cancer, and the primary mode of transmission is sexual contact. This provides context for the findings that younger age at first intercourse and an increasing number of lifetime sexual partners are both associated with an increased risk of developing cervical cancer. Pooled, individual, patient-level data from 12 cohort and case-control studies demonstrated statistically significantly increased risks of developing cervical cancer in women who were aged 17 years or younger at first intercourse, compared with women who were aged 21 years or older at first intercourse (relative risk [RR] for squamous cell cancer, 2.24; 95% confidence interval [CI], 2.11–2.38 and RR for adenocarcinoma, 2.06; 95% CI, 1.83–2.33). Similar findings were observed in women who had six or more lifetime sexual partners compared with women who had one lifetime sexual partner (RR for squamous cell cancer, 2.98; 95% CI, 2.62–3.40 and RR for adenocarcinoma, 2.64; 95% CI, 2.07–3.36).[11]

High parity

High parity has long been recognized as a risk factor for cervical cancer, but the relation of parity to HPV infection was uncertain. A meta-analysis of 25 epidemiologic studies, including 16,563 women with cervical cancer and 33,542 women without cervical cancer, showed that the number of full-term pregnancies was associated with increased risk, regardless of age at first pregnancy. This finding was also true if analyses were limited to patients with high-risk HPV infections (RR, 4.99; 95% CI, 3.49–7.13 for seven or more pregnancies vs. no pregnancies; linear trend test X2 = 30.69; P < .001).[12]

Long-term use of oral contraceptives

Long-term use of oral contraceptives has also been known to be associated with cervical cancer, but its relation to HPV infection was also uncertain. A pooled analysis of HPV-positive women from the studies described above was undertaken. Compared with women who have never used oral contraceptives, those who have used them for fewer than 5 years did not have an increased risk of cervical cancer (odds ratio [OR], 0.73; 95% CI, 0.52–1.03). The OR for women who used oral contraceptives for 5 to 9 years was 2.82 (95% CI, 1.46–5.42), and for 10 or more years, the OR was 4.03 (95% CI, 2.09–8.02).[13] A meta-analysis of 24 epidemiological studies confirmed the increased risk associated with oral contraceptives, which is proportionate to the duration of use. Risk decreases after cessation and returns to normal risk levels in 10 years.[14]

Cigarette smoke exposure

Cigarette smoking by women is associated with an increased risk of squamous cell carcinoma.[4,15,16] This risk increases with longer duration and intensity of smoking. The risk among smokers may be present with exposure to environmental tobacco smoke and may be as high as four times that of women who are nonsmokers who are not exposed to environmental smoking.[4] Case-control studies of women infected with HPV have examined the effect of various types and levels of tobacco exposure and found similar results.[16–18]

DES exposure

Diethylstilbestrol (DES) is a synthetic form of estrogen that was prescribed to pregnant women in the United States between 1940 and 1971 to prevent miscarriage and premature labor. DES is associated with a substantially increased risk of developing clear cell adenocarcinoma of the vagina and cervix among the daughters of women who used the drug during pregnancy (standardized incidence ratio, 24.23; 95% CI, 8.89–52.74); the risk persists as these women age into their 40s.[19] Despite the greatly elevated risk relative to the general population, this type of cancer is still rare; about one in 1,000 daughters exposed to DES will develop a clear cell adenocarcinoma.

DES exposure in utero is also associated with an increased risk of developing cervical dysplasia. An evaluation of three cohorts, including the Diethylstilbestrol Adenosis study, the Dieckmann study, and the Women’s Health Study, with long-term follow-up of more than 4,500 women exposed in utero to DES, found that 6.9% of exposed women developed grade II or higher CIN compared with 3.4% of nonexposed women (hazard ratio, 2.28; 95% CI, 1.59–3.27).[20]

Sexual abstinence

Nearly all cases of cervical cancer are associated with HPV infection, which is transmitted during sexual activity. Therefore, cervical cancer is seen more frequently in women with sexual activity at an early age and with multiple partners.[21] Lifetime abstinence from sexual activity is associated with a near-total reduction in the risk of developing cervical cancer.

HPV vaccination

Given the etiologic role of HPV in the pathogenesis of cervical neoplasia, vaccines to immunize against HPV infection offer a primary prevention strategy for cervical cancer. A quadrivalent (HPV-6, -11, -16, and -18) vaccine using a late protein L1 construct to induce antibody-mediated immunity was approved for use by the U.S. Food and Drug Administration in 2006; a bivalent (HPV-16, -18) vaccine was approved in 2009; and a vaccine targeting nine HPV types was approved in 2014. Vaccination during pregnancy has not been associated with adverse pregnancy outcomes.[22]

Persistent infection with oncogenic types of HPV, such as HPV-16 and HPV-18, is associated with the development of cervical cancer.[23] A vaccine to prevent HPV infection with oncogenic-type viruses has the potential to reduce the incidence of cervical cancer. A vaccine against HPV-16 using empty-viral capsids called virus-like particles (VLP) was developed and tested for efficacy in preventing persistent infection with HPV-16.

A multicenter, double-blind, placebo-controlled trial enrolled 2,391 women aged 16 to 23 years and randomly assigned them to receive either 40 µg of HPV-16 L1 VLP vaccine or placebo on day 1, at 2 months, and at 6 months. Papanicolaou (Pap) tests and genital samples for HPV-16 DNA were obtained on day 1, at 7 months, and every 6 months for 48 months. Colposcopy and cervical biopsies were obtained when clinically indicated at study exit. Serum HPV-16 antibody titers were obtained at study entry, at 7 months, and then every 6 months. A total of 1,505 women (755 receiving vaccine and 750 receiving placebo) completed all three vaccinations and had follow-up after month 7. After immunization, HPV titers peaked at month 7, declined through month 18, and then stabilized in months 30 through 48. There were no cases of CIN in the vaccine-treated women, but there were 12 cases in the placebo group (six CIN 2 and six CIN 3). HPV-16 infection that persisted for at least 4 months was seen in seven vaccine-treated women compared with 111 placebo-treated women.[24]

An international, double-blind, placebo-controlled trial of a bivalent HPV-16/HPV-18 VLP vaccine was performed in 1,113 women aged 15 to 25 years with normal cervical cytology who were seronegative for HPV-16, HPV-18, and 12 other oncogenic HPV types at enrollment. Women received either vaccine or placebo at 0, 1, and 6 months and were assessed by cervical cytology and self-obtained cervicovaginal samples for at least 18 months. A masked treatment-allocation follow-up study was performed for an additional 3 years, for a combined analysis of up to 6.4 years of follow-up. The 12-month persistent infection rate of HPV-16 or HPV-18 in an according-to-protocol cohort (i.e., women who received all three doses of vaccine or placebo on the correct schedule) was 0 of 401 women in the vaccine arm compared with 20 of 372 women in the placebo arm, with a vaccine efficacy of 100% (95% CI, 81.8–100). Diagnoses of CIN 2 or higher in a total vaccinated cohort (i.e., women who received at least one dose of vaccine or placebo) were 0 of 481 women in the vaccine arm compared with 9 of 470 women in the placebo arm, with a vaccine efficacy of 100% (95% CI, 51.3–100). Adverse events were similar in vaccinated and placebo-treated women. Neither analysis was intention-to-treat (ITT), making it difficult to know what the true vaccine efficacy for either virological or cytohistological endpoints would be in the routine clinical setting. Furthermore, cytohistological outcomes were reported only as composite endpoints (CIN 2+), making it impossible to distinguish the vaccine’s efficacy against invasive cervical cancer alone and potentially inflating the observed efficacy by including lesions with a relatively high probability (approximately 50% for CIN 2 [25]) of spontaneous regression.[26]

A quadrivalent vaccine (HPV types-6, -11, -16, and -18) was evaluated in a multinational, double-blind, randomized controlled trial of 17,622 women aged 15 to 26 years (FUTURE I and II).[27] Women received either the HPV vaccine or placebo at 0, 2, and 6 months; participants were assessed by clinical exam, Pap test, and HPV DNA testing for 4 or more years. Two analyses were reported. One group was considered to be HPV naive: negative to 14 HPV types. The second group was an ITT analysis, which approximates a sexually active population. The composite endpoint for cervical disease included the incidence of HPV-16/HPV-18–related, CIN 2, CIN 3, adenocarcinoma in situ, or invasive carcinoma. Outcomes were reported as follows:

This study also demonstrated decreased rates of abnormal Pap tests and subsequent diagnostic procedures. No cases of invasive cervical cancer were identified during the trial.

A 9-valent VLP vaccine was studied in another international randomized trial, which included 14,215 women. This new vaccine 9vHPV includes the four HPV types in the quadrivalent vaccine, qHPV (6, 11, 16, 18) and also 5 more oncogenic types (31, 33, 45, 52, 58). Sexually active women aged 16 to 26 years with fewer than five lifetime sexual partners received three intramuscular injections (day 1, month 2, and month 6) of either the qHPV vaccine or the 9vHPV vaccine. Women were evaluated every 6 months up to 5 years. The rate of high-grade cervical, vulvar, or vaginal disease was the same in both groups (14.0 per 1,000 person-years) because of pre-existing HPV infection, but the rate of disease related to HPV-31, -35, -45, -52, and -58 was lower in the 9vHPV vaccine group (0.1 vs. 1.6 per 1,000 person-years). Injection-site reactions were more common in the 9vHPV group.[28] Although not addressed in this study, the benefit of HPV vaccination is optimal in younger females before the onset of sexual activity.

All forms of the HPV vaccine are currently recommended by the Centers for Disease Control and Prevention (CDC) in the United States as a two-dose schedule at least 6 months apart for adolescents younger than 15 years. The current CDC recommendation for older individuals is to receive the original three-dose series. Recently, given issues of cost and adherence, there has been published data investigating whether similar vaccine efficacy could be obtainable using a reduced-dose schedule. A post hoc combined analysis of two phase III randomized controlled trials of the bivalent HPV vaccine (the Costa Rica Vaccine Trial and the PApilloma TRIal against Cancer In young Adults [PATRICIA] Trial) found that among women who were not HPV positive at enrollment for the specific virus type being studied, vaccine efficacy against either one-time incident detection of HPV 16/18 or incident infection that persisted at least 6 months was not statistically significantly different for those who received all three, two, or only one of the scheduled HPV vaccine doses (resulting from nonadherence or other factors) for up to 4 years of follow-up. Vaccine efficacy rates for persistent HPV 16/18 infection ranged from 89.1% (95% CI, 86.8%–91.0%) for three doses, to 89.7% (95% CI, 73.3%–99.8%) for two doses, to 96.6% (95% CI, 81.7%–99.8%) for one dose. To date, there are no randomized controlled trials that directly assess this clinical question.[29] A recent international study compared a two-dose schedule with a three-dose schedule in adolescents younger than 15 years who received the 9-valent HPV vaccine. The antibody response was noninferior in the two-dose schedule, supporting the current recommendation that two doses are sufficient for this age group.[30]

On the basis of their mechanism of action, L1/2 HPV vaccines do not appear to impact pre-existing infections. The FUTURE II trial demonstrated a markedly lower vaccine efficacy rate in the total randomized study population, which included individuals who were positive for HPV at baseline, compared with the per-protocol population (44% for lesions associated with HPV-16 or HPV-18, and 17% for lesions associated with any HPV type vs. 98%; refer to Table 1 above).[27] Additionally, an intermediate analysis of a randomized controlled trial primarily evaluating the efficacy of the HPV-16/18 vaccine in preventing infection found no effect on viral clearance rates in women aged 18 to 25 years who were positive at the time of study enrollment.[31]

The type-specific vaccines, if successful in preventing invasive cancer, will offer protection for only a subset of cases, the proportion of which will vary worldwide.[32] Using data from a multicenter case-control study conducted in 25 countries, it was estimated that a vaccine containing the seven most common HPV types could prevent 87% of cervical cancers worldwide. A vaccine with the two most common strains, HPV-16 and HPV-18, would prevent 71% of cervical cancers worldwide.[32]

A study of cervical HPV DNA among 202 Australian women aged 18 to 24 years who were sampled between 2005 and 2007 before implementation of a national quadrivalent prophylactic HPV vaccine program compared the results with a matched group of 1,058 women who were sampled in the postvaccination era (2010–2012). This study found an adjusted prevalence ratio among fully vaccinated women of 0.07 (95% CI, 0.04–0.14; P < .0001) for vaccine-related HPV types and a smaller but statistically significant magnitude of protection of 0.65 (95% CI, 0.43–0.96; P < .03) among unvaccinated women, suggesting herd immunity (protection of unvaccinated individuals).[33] These data strengthen previous results that suggest herd immunity in this population manifested as a reduction in genital warts among heterosexual men, a group that includes sexual partners of vaccinated women.[34] Data also suggest cross protection against carcinogenic types that are not directly targeted by the quadrivalent vaccine but are included in the new nonvalent HPV vaccine.[33]

There are data that explore the impact of national HPV vaccination programs and report on vaccine effectiveness. In England, 15,459 residual genital specimens from women aged 16 to 24 years, collected for chlamydia screening between 2010 and 2016, were utilized for national HPV surveillance.[35] In this study, vaccine effectiveness for HPV-16/HPV-18 was 82% (95% CI, 60.6%–91.8%) for women who were vaccinated before age 15 years. Within the younger age groups, the prevalence of HPV-16/HPV-18 significantly decreased within the postvaccination period between 2010 and 2011 to 2016 from 8.2% to 1.6% in 16 to 18 year olds and from 14.0% to 1.6% in 19 to 21 year olds (compared with 17.6% and 16.9% in the prevaccination era).[35]

Use of barrier method during sexual intercourse

Barrier methods of contraception are associated with a reduced incidence of SIL presumptively secondary to protection from sexually transmitted disease.[36,37] The effectiveness of condom use for the prevention of HPV infections has been evaluated in a prospective study of women aged 18 to 22 years who were virgins.[38] The number of vulvovaginal HPV infections was reduced with consistent condom use, and HPV infection rate was 37.8 infections per 100 patient-years among women whose partners used condoms 100% of the time in the 8 months before testing, compared with 89.3 infections per 100 patient-years among women whose partners used condoms less than 5% of the time (P trend = .005). No cervical SIL were detected among women reporting 100% condom use by their partner.[38]